Lease or buy? How to decide on a service robot without a spreadsheet full of guesses
The monthly lease rate is the least important number in the comparison. Utilisation, residual value and who carries the technology risk decide the answer, and they can be estimated before you ask for a price.
Start with trips per day, not with price
A service robot is a labour substitution with a capital cost. The only honest way to compare it with the alternative is to count what it replaces. For a hotel, that is the number of delivery trips per day — room service, amenities, laundry, minibar restock — multiplied by the minutes each trip takes a person, multiplied by the loaded hourly cost of that person in your market.
Below roughly 20 trips a day, a single robot is usually idle enough that the payback period stretches past the warranty, and the project quietly becomes a marketing decision rather than an operational one. That is a legitimate reason to buy a robot. It is a bad reason to believe a spreadsheet that says otherwise.
What leasing actually buys you
A lease is not cheaper. It is a different distribution of risk. In a three-year lease you are paying a finance company for the right to hand the machine back when the technology is obsolete, and you are paying a premium for that right. What you get in return:
- The capital stays on the operator's balance sheet as an operating cost, which matters for asset-light management contracts.
- Technology risk transfers to the lessor if the platform is superseded mid-term.
- Maintenance is usually bundled, so the cost is predictable per month rather than lumpy.
- Exit is defined. If the robot does not fit the operation, you are not reselling a used machine into a thin market.
What you give up: total cost is higher over the asset's life, you usually cannot modify the machine, and end-of-term purchase options are priced in the lessor's favour.
What buying actually buys you
Ownership wins when utilisation is high and the deployment is stable. A hospital logistics route that runs 60 trips a day for five years is a purchase, not a lease. Ownership also wins when you intend to customise — compartment layout, branding, interface language, integration with your PMS or WMS — because a lessor will not accept a modified asset as readily as you will.
The hidden cost of ownership is the residual value assumption. Service robots do not hold value like vehicles. There is no Blue Book, the secondary market is thin, and a three-year-old platform with a discontinued lidar is worth parts. If your business case depends on resale, it is not a business case.
The decision rule we actually use
| Situation | Usually better | Reason |
|---|---|---|
| Under 20 trips a day, one site | Lease, or do not buy | Payback exceeds the warranty period at realistic labour costs |
| 20 to 60 trips a day, stable route | Buy | Utilisation absorbs the capital cost inside 24 to 30 months |
| Multiple sites, standardised deployment | Buy, negotiate volume | Per-unit price and spare-parts pooling both improve |
| Custom compartment or brand integration | Buy | A lessor will not finance a modified asset |
| Pilot with an unclear outcome | Lease short, or rent | Three months of data beats a three-year commitment |
The question nobody asks
Who empties it? A robot that carries 30 kg across four compartments still needs a person to load and unload it, and that person is usually on the night shift, which is the shift you were trying to protect. If the answer is "the same person who used to do the deliveries", you have bought a machine that makes one employee slightly faster, not one that removes a role. That is a real benefit. It is just not the benefit in the proposal.
Ask for the numbers on your own trip count and your own labour cost. Any supplier who quotes a payback period without asking those two questions is quoting a brochure.
Tell us the site. We will tell you what fits.
Send the corridor width, the lift door, the floor plan and the number of trips per day. You get a model recommendation and a written quotation within one working day — or an honest answer that a robot is not the right purchase for that route.
Ask for a quotation